Card-linked loyalty, and who ends up holding your data
Yoco has switched on card-linked loyalty for small business. There is no app to download and no integration to scope: the customer joins with a phone number, gets their balance over WhatsApp and redeems at the counter. The cost is not the discount. It is that your customer history now accumulates inside your payment processor.
Card linked loyalty for small business in South Africa has been promised for a decade and has died every time on friction. This version removes the friction. The question it raises is a different one.
What actually shipped
Yoco Loyalty was announced at Yoco Next and written up by The Open Letter. It is card linked, meaning the customer's bank card becomes the loyalty card, and it is switched on from the Plus plan rather than commissioned separately. Staff have nothing new to run.
The precedent was Platō, which The Open Letter reported last year as the first South African business to make loyalty card linked, using Yoyo on Yoco terminals with 5% cashback landing automatically. Within three weeks the Platō app accounted for over 30% of in store transactions. It was still an app, and it still needed integration.
Market context is not the constraint here. Truth and BrandMapp research, cited in the same piece, puts 85% of economically active South Africans in loyalty programmes, averaging 10.4 each, double the 2015 figure. The country will join. It always joins.
The numbers underneath, and where they come from
The Open Letter reports that in Q1 2026, R4.9 billion in transactions from repeat customers moved through Yoco's food and beverage network. One in three of those customers returned at least twice a month, and about half of all revenue came from roughly 38% of the customer base.
Those are newsletter figures rather than a Yoco publication, and they should be read as reported rather than as audited. If they are close to right, they describe chain grade customer concentration sitting inside thousands of businesses that have never been able to see it.
The same piece reports results at two merchants. At Suur, loyalty members spent 2.2 times more than non members over two months, driven by frequency rather than basket size, at 4.4 visits a month against 2. Brewsky Coffee passed 669 members in weeks and got back about two hours a week of stamp card reconciliation. Both are the merchant's own numbers, reported second hand.
What it costs, in rands
The reward rate is yours to set, and the reward comes off margin. Platō's version costs 5% of every cup.
Work it on stated assumptions and substitute your own. A shop turning over R400 000 a month on card, where roughly half of that revenue comes from the regulars a loyalty programme is designed to reach, is rewarding about R200 000 of turnover. At 5% that is R10 000 a month, or R120 000 a year, against a frequency lift you will not be able to measure for at least a quarter.
That is an affordable experiment for a business with a gross margin above 60%, and a serious decision for one below 40%. The point is that it is a decision with a number attached, which is more than most loyalty pitches offer.
What blocks it here
Two things, and both are particular to this market.
Card linked only recognises the customer tapping the same card. A business where a meaningful share of takings is cash will be measuring the card half of its customer base and calling it the business. That is not a small distortion in most of South Africa, and it skews exactly the segment analysis the programme is meant to produce.
The second is the switching cost, and it compounds. The customer history builds up with the payment processor rather than in anything you own, which makes changing processors more expensive every month you run the programme. A year in, the loyalty data is a reason to stay that has nothing to do with the pricing of the terminal.
There is a POPIA question sitting alongside it that nobody in the launch coverage will ask. A phone number tied to a purchase history is personal information. Establish in writing who is the responsible party and who is the operator for that customer list, and what happens to it when the relationship ends.
What this is worth depends on what it joins to
A customer list at your processor and an invoicing system that has never met it gives you a marketing list, not a customer record. The value in knowing that 38% of customers make half your revenue only arrives when that fact can reach your ordering, your staffing roster and your sales records.
Connecting a payment platform's export to the systems a business already runs, rather than replacing anything, is the ordinary work Zorah does. It is also the difference between a loyalty programme that pays for itself and one that is a monthly discount with good branding.
What to do on Monday
Ask your processor two questions in writing, before you switch anything on. Can you export the full customer list and transaction history, in what format, and does that export survive you leaving.
Then count last month's cash takings as a share of total turnover. That percentage is the part of your customer base this programme will never see, and it decides whether the data is worth anything at all.
You can read the product page after you have both answers, not before.